Statistical Methods and Graphical Displays for Analyzing How the Association Between Two Qualitative Variables Differs Among Countries, Among Groups, Or Over Time: A Modified Regression-Type Approach

Many interesting sociological questions pertain to how the association between two variables depends on a third variable. In sociological applications, the third variable often pertains to countries, to subgroups of a population, or to time periods. We propose a regression-type approach that specifies that the log-odds- ratios that describe the two-way association of interest are a linear function of latent scores for the third variable. Additive and multiplicative models currently in use by researchers are special cases of the regression-type model. To illustrate the utility of the regression-type approach, we apply this approach to analyze (1) data on occupational mobility in the United States, Britain, and Japan (comparing mobility in these countries) and (2) data on the association between religion and voting behavior in U.S. presidential elections from 1968 to 1992 (comparing this association in the different elections). We also introduce here graphical displays that can be used to obtain worthwhile information about goodness-of-fit and to aid in substantive interpretation.     

Explaining the Flight of Cupid’s Arrow: A Spatial Random Utility Model of Partner Choice

Spatial homogamy may be defined as follows: anyone may be attracted to anyone else, but near candidates are more attractive than distant candidates. In this article, we propose a model of partner choice, where homogamy is defined in terms of spatial, demographic, socioeconomic and cultural similarity. A spatial choice model using random utility theory is formulated, taking into account a relaxation of the independence from the irrelevant alternatives property, as spatial alternatives are not independent of one another. We model partner choice given the characteristics of the chosen partner and a choice set of alternatives, using unique micro data on all new cohabiters in the Netherlands, linked to other relevant data sets. The model takes the spatial locations of potential candidates within a choice set into account, including an indicator for the spatial similarity between alternatives. We find that spatial homogamy is a vital component of partner matching, aside from and adding to the spatial effects in demographic, socioeconomic and cultural homogamy. Given a choice set of partners, the highest likelihood of a match occurs with a person who is born and lives near by, who is close in age, is in the same life stage and has the same marital status, who has the same educational and income level and the same labour market status, who speaks the same dialect and lives in a culturally similar residential area. The distance effect is most pronounced for those individuals with lower levels of education and those living in rural areas.     

Mathematical programming: Turing completeness and applications to software analysis

Mathematical programming is Turing complete, and can be used as a general-purpose declarative language. We present a new constructive proof of this fact, and showcase its usefulness by discussing an application to finding the hardest input of any given program running on a Minsky Register Machine. We also discuss an application of mathematical programming to software verification obtained by relaxing one of the properties of Turing complete languages.     

Predicting 1-year change in body mass index among college students

OBJECTIVE: Despite beliefs about weight gain in college, few researchers have evaluated this phenomenon. PARTICIPANTS: Participants were 18- to 31-year-old students at a midwestern university. The dependent variable was body mass index (BMI) change. METHODS: The authors extracted predictor variables from a Health Risk Appraisal. These included clinical, medical history, medical usage, medications, pain or chronic conditions, perceptual measures, and behavioral factors. The authors performed an ordinal regression technique separately by gender. RESULTS: No predictors were significant for men. Women in the BMI gain group were (1) more likely to consume alcohol, use maladaptive coping behaviors, eat foods low in fiber, and consume caffeine; and (2) less likely to be stress-free, to eat cruciferous vegetables, and to refrain from eating high-cholesterol foods. CONCLUSIONS: The lack of research on predictors of and interventions for reducing BMI gain among college students warrants more research.     

FDI’s Impact on Transitional Countries, Serbia as a Rational Choice: The FIAT-ZASTAVA Case

The privatization process and foreign direct investments (FDI) are among the most important economic issues in transitional countries. FDI has enormous influence on transitional countries that need serious structural changes. In this paper we will study FIAT’s decision to invest in the Serbian automotive industry. We will try to rationalize FIAT’s decision because Serbian car manufacturer ZASTAVA was not the only alternative for FIAT to invest in the Balkans. In order to justify FIAT’s decision we will apply the Mullins model of horizontal and vertical FDI, examine the importance of the cooperating history between two car manufacturers and compare macroeconomic conditions and Global Competitiveness of Serbia and Romania, which was the main alternative for FIAT. Finally, we will also try to explain why FIAT decided to announce the investment at the end of political campaign in Serbia when pools gave anti-European parties higher chances for victory, instead of waiting to see the outcome of the elections.

Psychophysiological Reactivity in Child Sexual Abuse

Sexual abuse has physiological and emotional implications. The purpose of this study is to evaluate the neurobiological sequels of childhood sexual trauma by monitoring physiological variables among sexually abused girls and women compared to controls. We assessed posttrauma and traumatic life events of 35 females sexually abused in their childhood (age range 7–51 years) and 25 control females (age range 7–54 years). Electroencephalography, frontalis electromyography, electrodermal activity, and heart rate parameters were recorded while watching sets of pictures representing neutral and trauma-suggestive stimuli. A minority of participants met the Diagnostic and Statistical Manual of Mental Disorders criteria for post-traumatic stress disorder. Abused females displayed significant elevations in heart rate, electromyography, and electroencephalography while viewing allusive stimuli and elevated heart rate while viewing neutral stimuli. The dysfunctional regulation of the physiological stress system associated with child sexual abuse may endanger the victims with various stress and anxiety disorders.